RK-1235's picture
Add new SentenceTransformer model
795ee69 verified
---
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:801
- loss:MultipleNegativesRankingLoss
base_model: BAAI/bge-base-en
widget:
- source_sentence: Who signed the case report and provided their signature at the
end of the document?
sentences:
- 'First Information Report: The arrest of 30 quarts of country liquor occurred
under Section 60 of the Excise Act by a team at Nawabad Police Station, Jhansi
district. Today on the date 24/02/24, I, Mr. Brajesh Kumar Bhargava, along with
Constable Lokendra Singh, filed report number 31 at 10:04 AM. We were patrolling
the Binavar Care Area when we received special intelligence from the senior officer
that an individual carrying white liquor was standing near the toilet behind the
bus stand at Haveli Market. If acted quickly, he could be apprehended. Following
the senior officer''s information, I, Mr. May Hamrahi, accompanied by Constable
Lokendra Singh, proceeded towards the bus stand in Ambedkar Nagar and moved towards
the toilet inside Tala Pura. The senior officer signaled from a distance and returned.
About 15 steps before the toilet, a person was seen standing with a white-colored
bottle in his hand. As soon as the police approached him, he tried to run away
but was caught on the'
- 'upon reaching the police station. Signatures: H/O Upendra Signature Navratan
Singh U/N Signature Bhim Singh Signature English Unreadable SI 18/3/24 (Navin
Tomar) U/N Police Station No. 113 Gautambuddhanagar Note: A copy of the case was
given to the accused Mizruul above. Fingerprint Accused LTI Mizruul Signature
English Unreadable SI 18/3/24 (Navin Tomar) U/N Police Station No. 113 Gautambuddhanagar
5 NC R.B (N.C.R.B) I.L.F.-I (Unified Investigation Form -1)'
- I.l.F.-I (Unified Investigation Form -1) verified and found true.
- source_sentence: Who prepared and signed the arrest report?
sentences:
- 'First Information Report: Counterfeit letter service officer Shri Prabhari Inspector
Sir Thana Khesra District Si. City Sir, respectfully informed that I am the petitioner
Anurag Shukla SLO Raakesh Shukla village Mirwapur PO Beluakh Si. city resident
today dated 28/02/24 my father Shri Raakesh Shukla Marvatia Market was returning
home on a motorbike when around 5 PM in Basdia village near Pinto''s tea shop
as they reached there an old Ranjis regarding Raju alias Tira son Ramchandra alias
Kallu Ajay son Avadhash Sanee Devl son Rajesh and some other people gathered together
started beating my father with abuses calling him mother sister with sticks and
slaps which caused my father to fall injured some people intervened and saved
my father then the above-mentioned people left after threatening to kill my father
we took our father to CHC Beloha where the doctor referred him to Naugadh due
to severe injuries where his treatment is ongoing. Therefore, it is requested
to Shri Man Ji to take'
- 'First Information Report: Counterfeit currency—seizure and search operation.
Cash seizure of 380 rupees, one pen, one person arrested, charged under Section
13 G ACT, Thana Sadar Bazaar, Shahjahanpur District, today''s date 20/3/2024.
Officer ID 195, Devendra Kumar, along with colleague Officer ID 1774, Somveer,
accompanied by informant, left from Haja police station and stopped at Ashfaqnagar
area for routine checks, crime prevention, and peacekeeping measures. When the
police approached near Baliji Temple, they received special intelligence from
an informant that a person standing near Chamakni Police Station was involved
in counterfeit currency transactions and could be caught quickly. The police believed
the informant and conducted searches on each other to ensure no one had criminal
material. Accompanied by the informant, they proceeded to the location indicated
by the informant. Upon reaching Sharab Bhatti Sahibaznagar Road, the informant
pointed out from a distance that someone'
- and giving testimony, and they left without providing their names due to fear.
Throughout the entire process, orders and guidelines of the Honorable Supreme
Court and the National Human Rights Commission were fully followed. The arrest
notice will be given to the concerned parties through appropriate means at the
police station. This report was prepared by speaking and writing by me, Pradeep,
and written by Ka 1952 Susant, read out and signed by all concerned
- source_sentence: Where did the accident involving Ramesh Singh Kushwaha occur relative
to the police station?
sentences:
- 'First Information Report: The complaint is in Hindi. Mr. Thana Adhyaksh Mahodaya
Thana Birnora Janpad Gajipur Mahodaya is respectfully requested that the applicant
Rajkumar Singh Kushwaha son of Chandrajit Singh Kushwaha, originally from village
Araji Odasun, Thana Birnora, Janpad Gajipur, on today''s date 12.03.2024, around
7:30 AM, my brother Ramesh Singh son of Chandrajit was going to check the fields
from Gajipur Road when, about 200 meters from the police station, near Singh Kushwaha''s
house, in the western direction towards Ajmer, vehicle number UPSOABSO99 carelessly
hit Ramesh Singh Kushwaha''s rider from behind with great force, resulting in
my brother Ramesh Singh Kushwaha dying on the spot. Therefore, it is requested
that Mr. Ji kindly write a report (register) and take necessary action. Applicant''s
signature Rajkumar Singh Kushwaha son of Chandrajit Singh Kushwaha, village Araji
Odasun, Thana Birnora, Janpad Gajipur, mobile 9532127579, date 12.03.2024. Note:
I, Kanu Amit Kumar'
- Kumar Praja Pit 4 NC RB (NCRB) I.l.F.- (Unified Investigation Form-l)
- attached with screws. The hammer is made of iron, the total length being 2 cm
(1 inch). The trigger guard is made of iron with a screw for opening and closing
attached to an iron plate on the other side. The seized firearm is a brass cartridge
marked KFSMM on its base. This act falls under the chargeable offense under Section
3/25 A Act. At around 10:10 am, while informing about the arrest, the orders and
guidelines of the Human Rights Commission and the Supreme Court were followed,
and Himanshu Singh son of Veerendra Singh, residing in Paradhan, Kotwali town,
Maun district, was taken into custody under Section 3/25 A Act. One motorcycle
vehicle numbered UPS4 AB7344 was seized from the scene, and when asked for registration,
he could not show any, which is an offense under 207 M.V. Act. A report has been
filed under 207 MV Act. The seized illegal firearm and cartridge were placed in
a transparent plastic box brought by Constable Ka0 Manu Kumar and sealed with
a sample seal. The case
- source_sentence: How much money was given to Rajkumar Kanaujia on February 2, 2019?
sentences:
- 'cost Rupees 5 lakhs. Thus, under the pretext of getting the applicant and his
brother Mo Arju''s government jobs in the Panchayat Raj Department, a total of
Rupees 10 lakhs was given to the aforementioned Rajkumar Kanaujia. The details
are as follows: On February 2, 2019, Rupees 10,000 cash; on February 6, 2019,
Rupees 40,000 cash; on August 17, 2019, Rupees 1,00,000 cash; on October 17, 2019,
Rupees 49,000 cash; on July 23, 2020, Rupees 1,00,000 cash; on March 8, 2021,
Rupees 50,000 cash; on March 26, 2021, Rupees 50,000 cash. The applicant gave
this money directly to Rajkumar Kanaujia and withdrew from his sister-in-law''s
account. On February 27, 2020, Rupees 22,000 cash; on September 18, 2020, through
NEFT, Rupees 1,00,000/-. Check No. 908768, Total Rupees 5,21,000/-. On October
22, 2021, Rupees 1,00,000 was given to Puran''s wife Sony, who was sent by Rajkumar
Kanaujia. The remaining Rupees 3,79,000/- was given in cash by the applicant to
Rajkumar Kanaujia after borrowing from'
- 'First Information Report: The complaint is in Hindi. Mr. Thana Adhyaksh Mahodaya
Thana Birnora Janpad Gajipur Mahodaya is respectfully requested that the applicant
Rajkumar Singh Kushwaha son of Chandrajit Singh Kushwaha, originally from village
Araji Odasun, Thana Birnora, Janpad Gajipur, on today''s date 12.03.2024, around
7:30 AM, my brother Ramesh Singh son of Chandrajit was going to check the fields
from Gajipur Road when, about 200 meters from the police station, near Singh Kushwaha''s
house, in the western direction towards Ajmer, vehicle number UPSOABSO99 carelessly
hit Ramesh Singh Kushwaha''s rider from behind with great force, resulting in
my brother Ramesh Singh Kushwaha dying on the spot. Therefore, it is requested
that Mr. Ji kindly write a report (register) and take necessary action. Applicant''s
signature Rajkumar Singh Kushwaha son of Chandrajit Singh Kushwaha, village Araji
Odasun, Thana Birnora, Janpad Gajipur, mobile 9532127579, date 12.03.2024. Note:
I, Kanu Amit Kumar'
- 'lid was opened. When asked for a license to carry raw liquor, he could not show
it and requested forgiveness for his crime. The accused was taken to the custody
of the police around 16:35. The lid of the seized bottle was sealed and stamped
according to the Supreme Court order and instructions. No witnesses were available
due to various reasons, and they left without giving their names and addresses.
The incident report was written by me, Constable Shivdatt, and signed by my colleague
Constable Somveer. The arrest information will be communicated to the public through
appropriate channels upon reaching the police station. The signature is illegible
in English by Constable Shivdatt, serial number 2175, Shahjahanpur Sadar Bazaar
Police Station, dated 23.01.2024. The signature is legible in Hindi by Constable
Somveer. A carbon copy of the report was given to the accused for acknowledgment.
Note: I, Amarjeet Kumar, the main writer, certify that the copy of the report
has been fed word by'
- source_sentence: Under which section of the law was Rahul Shishupal apprehended?
sentences:
- Mann son Saudan residing Kanhor Thana Pahadi Bharatpur Rajasthan, 5. Mustak son
Hariya residing Ghagwadi Thana Kathwada Bharatpur Rajasthan, 6. Ikram son Iliaas
residing Jamat Thana Punhana Noonu Mevata Haryana, 7. Nasir son Raju residing
Bisru Thana Punhana Noonu Mevata Haryana, 8. Sahun son Rahmuddin residing Dhaulpur
Thana Dhaulpur Faridabad Haryana, 9. Hakku alias Hakmu alias Hakmuddin son Nasru
alias Nasrudeen residing Bichhor Thana Bichhor Noonu Mevata Haryana, 10. Kala
alias Kala Minister son Kanura alias Kalua residing Kanhor Thana Pahadi Bharatpur
Rajasthan are active members. The main activity of this gang is robbery, looting,
and illegal smuggling which involves illegal money making, earning money to support
themselves and their families, causing considerable fear and terror in the area.
Due to the fear of this gang, no person from the public is willing to give testimony
or write reports against them. These are habitual criminals described under Chapter
17 of the IPC. It is
- 2023, Mr. Chakrapani Katiyar son of Ishwardayal Katiyar Balaji Hospital and SCAASTAKA
Diagnostic Center Hardoi reported that the gang leader and gang members had created
a fake web portal and spread false news through various news channels to the public
and demanded 50,000 rupees per month from hospital staff and forcibly took 5,000
rupees from hospital staff on May 25, 2023. A case FIR No. 360/2023 under sections
34/386/506 IPC was registered at Kotwali City Police Station. After investigation,
based on evidence, a charge sheet was sent to the court against the gang leader
and members which is pending in the court. On May 28, 2023, Mr. Sudhanshu Mishra
son of OP Mishra residing Bahra Saudagar Lucknow Road Thana Kotwali City Hardoi
reported that he was beaten, abused, and threatened with death and demanded 5,000
rupees by the gang leader DD Shukla and gang members Mohit Mishra, Ankrit Mishra,
and their other accomplices. A case FIR No. 35
- raw liquor near the temple in the forest. The informant then left. We hid our
two motorcycles and approached the person stealthily. When we reached the person,
we noticed a large crowd around him. Seeing us, the crowd dispersed. We apprehended
the person selling liquor with a show of force. After questioning the apprehended
individual about his name and address and conducting a search, he identified himself
as Rahul Shishupal, aged approximately 28 years, from Nanakganj Jhala village,
Kotwali Police Station, Hardeo. During the search, a plastic bottle containing
approximately 10 liters of raw liquor was seized. Opening the lid of the bottle,
we smelled a pungent odor of raw liquor. The act of the apprehended individual
falls under Section 60 ex.act. The accused was informed of his offense under Section
60 ex.act. When asked for authorization documents regarding the possession and
sale of illegal raw liquor, he was unable to produce them and requested forgiveness
for his mistake. The
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@3
- cosine_recall@5
- cosine_recall@10
- cosine_ndcg@10
- cosine_mrr@10
- cosine_map@100
model-index:
- name: SentenceTransformer based on BAAI/bge-base-en
results:
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 1024
type: dim_1024
metrics:
- type: cosine_accuracy@1
value: 0.3888888888888889
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.6823529411764706
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.7300653594771241
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.7973856209150327
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.3888888888888889
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.22745098039215686
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.1460130718954248
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.07973856209150326
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.3888888888888889
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.6823529411764706
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.7300653594771241
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.7973856209150327
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.6081898113831302
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.5461826953003428
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.5512418200825733
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 768
type: dim_768
metrics:
- type: cosine_accuracy@1
value: 0.3888888888888889
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.6823529411764706
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.7300653594771241
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.7973856209150327
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.3888888888888889
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.22745098039215686
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.1460130718954248
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.07973856209150326
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.3888888888888889
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.6823529411764706
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.7300653594771241
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.7973856209150327
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.6081898113831302
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.5461826953003428
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.5512418200825733
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 512
type: dim_512
metrics:
- type: cosine_accuracy@1
value: 0.3843137254901961
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.6718954248366014
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.7287581699346405
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.7875816993464052
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.3843137254901961
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.2239651416122004
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.14575163398692811
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.0787581699346405
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.3843137254901961
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.6718954248366014
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.7287581699346405
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.7875816993464052
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.6013957196390363
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.5402334267040153
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.5455156504447409
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 256
type: dim_256
metrics:
- type: cosine_accuracy@1
value: 0.3718954248366013
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.6431372549019608
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.7026143790849673
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.7679738562091504
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.3718954248366013
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.21437908496732025
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.14052287581699346
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.07679738562091502
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.3718954248366013
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.6431372549019608
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.7026143790849673
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.7679738562091504
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.5813509127848285
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.5204315800394235
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.5256057776792838
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 128
type: dim_128
metrics:
- type: cosine_accuracy@1
value: 0.3366013071895425
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.5915032679738562
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.6601307189542484
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.7169934640522876
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.3366013071895425
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.1971677559912854
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.13202614379084965
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.07169934640522875
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.3366013071895425
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.5915032679738562
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.6601307189542484
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.7169934640522876
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.5365811801879968
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.47775236020334133
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.4842596807377994
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 64
type: dim_64
metrics:
- type: cosine_accuracy@1
value: 0.2718954248366013
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.4849673202614379
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.5496732026143791
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.6209150326797386
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.2718954248366013
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.1616557734204793
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.10993464052287581
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.06209150326797386
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.2718954248366013
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.4849673202614379
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.5496732026143791
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.6209150326797386
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.4494736762212422
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.39436922917315087
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.4024049717422007
name: Cosine Map@100
---
# SentenceTransformer based on BAAI/bge-base-en
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-base-en](https://huggingface.co/BAAI/bge-base-en) on the json dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [BAAI/bge-base-en](https://huggingface.co/BAAI/bge-base-en) <!-- at revision b737bf5dcc6ee8bdc530531266b4804a5d77b5d8 -->
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 768 dimensions
- **Similarity Function:** Cosine Similarity
- **Training Dataset:**
- json
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("RK-1235/bge-base-FIR-matryoshka-BASELINE-10epo-triplet")
# Run inference
sentences = [
'Under which section of the law was Rahul Shishupal apprehended?',
'raw liquor near the temple in the forest. The informant then left. We hid our two motorcycles and approached the person stealthily. When we reached the person, we noticed a large crowd around him. Seeing us, the crowd dispersed. We apprehended the person selling liquor with a show of force. After questioning the apprehended individual about his name and address and conducting a search, he identified himself as Rahul Shishupal, aged approximately 28 years, from Nanakganj Jhala village, Kotwali Police Station, Hardeo. During the search, a plastic bottle containing approximately 10 liters of raw liquor was seized. Opening the lid of the bottle, we smelled a pungent odor of raw liquor. The act of the apprehended individual falls under Section 60 ex.act. The accused was informed of his offense under Section 60 ex.act. When asked for authorization documents regarding the possession and sale of illegal raw liquor, he was unable to produce them and requested forgiveness for his mistake. The',
'Mann son Saudan residing Kanhor Thana Pahadi Bharatpur Rajasthan, 5. Mustak son Hariya residing Ghagwadi Thana Kathwada Bharatpur Rajasthan, 6. Ikram son Iliaas residing Jamat Thana Punhana Noonu Mevata Haryana, 7. Nasir son Raju residing Bisru Thana Punhana Noonu Mevata Haryana, 8. Sahun son Rahmuddin residing Dhaulpur Thana Dhaulpur Faridabad Haryana, 9. Hakku alias Hakmu alias Hakmuddin son Nasru alias Nasrudeen residing Bichhor Thana Bichhor Noonu Mevata Haryana, 10. Kala alias Kala Minister son Kanura alias Kalua residing Kanhor Thana Pahadi Bharatpur Rajasthan are active members. The main activity of this gang is robbery, looting, and illegal smuggling which involves illegal money making, earning money to support themselves and their families, causing considerable fear and terror in the area. Due to the fear of this gang, no person from the public is willing to give testimony or write reports against them. These are habitual criminals described under Chapter 17 of the IPC. It is',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Information Retrieval
* Dataset: `dim_1024`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) with these parameters:
```json
{
"truncate_dim": 1024
}
```
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.3889 |
| cosine_accuracy@3 | 0.6824 |
| cosine_accuracy@5 | 0.7301 |
| cosine_accuracy@10 | 0.7974 |
| cosine_precision@1 | 0.3889 |
| cosine_precision@3 | 0.2275 |
| cosine_precision@5 | 0.146 |
| cosine_precision@10 | 0.0797 |
| cosine_recall@1 | 0.3889 |
| cosine_recall@3 | 0.6824 |
| cosine_recall@5 | 0.7301 |
| cosine_recall@10 | 0.7974 |
| **cosine_ndcg@10** | **0.6082** |
| cosine_mrr@10 | 0.5462 |
| cosine_map@100 | 0.5512 |
#### Information Retrieval
* Dataset: `dim_768`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) with these parameters:
```json
{
"truncate_dim": 768
}
```
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.3889 |
| cosine_accuracy@3 | 0.6824 |
| cosine_accuracy@5 | 0.7301 |
| cosine_accuracy@10 | 0.7974 |
| cosine_precision@1 | 0.3889 |
| cosine_precision@3 | 0.2275 |
| cosine_precision@5 | 0.146 |
| cosine_precision@10 | 0.0797 |
| cosine_recall@1 | 0.3889 |
| cosine_recall@3 | 0.6824 |
| cosine_recall@5 | 0.7301 |
| cosine_recall@10 | 0.7974 |
| **cosine_ndcg@10** | **0.6082** |
| cosine_mrr@10 | 0.5462 |
| cosine_map@100 | 0.5512 |
#### Information Retrieval
* Dataset: `dim_512`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) with these parameters:
```json
{
"truncate_dim": 512
}
```
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.3843 |
| cosine_accuracy@3 | 0.6719 |
| cosine_accuracy@5 | 0.7288 |
| cosine_accuracy@10 | 0.7876 |
| cosine_precision@1 | 0.3843 |
| cosine_precision@3 | 0.224 |
| cosine_precision@5 | 0.1458 |
| cosine_precision@10 | 0.0788 |
| cosine_recall@1 | 0.3843 |
| cosine_recall@3 | 0.6719 |
| cosine_recall@5 | 0.7288 |
| cosine_recall@10 | 0.7876 |
| **cosine_ndcg@10** | **0.6014** |
| cosine_mrr@10 | 0.5402 |
| cosine_map@100 | 0.5455 |
#### Information Retrieval
* Dataset: `dim_256`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) with these parameters:
```json
{
"truncate_dim": 256
}
```
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.3719 |
| cosine_accuracy@3 | 0.6431 |
| cosine_accuracy@5 | 0.7026 |
| cosine_accuracy@10 | 0.768 |
| cosine_precision@1 | 0.3719 |
| cosine_precision@3 | 0.2144 |
| cosine_precision@5 | 0.1405 |
| cosine_precision@10 | 0.0768 |
| cosine_recall@1 | 0.3719 |
| cosine_recall@3 | 0.6431 |
| cosine_recall@5 | 0.7026 |
| cosine_recall@10 | 0.768 |
| **cosine_ndcg@10** | **0.5814** |
| cosine_mrr@10 | 0.5204 |
| cosine_map@100 | 0.5256 |
#### Information Retrieval
* Dataset: `dim_128`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) with these parameters:
```json
{
"truncate_dim": 128
}
```
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.3366 |
| cosine_accuracy@3 | 0.5915 |
| cosine_accuracy@5 | 0.6601 |
| cosine_accuracy@10 | 0.717 |
| cosine_precision@1 | 0.3366 |
| cosine_precision@3 | 0.1972 |
| cosine_precision@5 | 0.132 |
| cosine_precision@10 | 0.0717 |
| cosine_recall@1 | 0.3366 |
| cosine_recall@3 | 0.5915 |
| cosine_recall@5 | 0.6601 |
| cosine_recall@10 | 0.717 |
| **cosine_ndcg@10** | **0.5366** |
| cosine_mrr@10 | 0.4778 |
| cosine_map@100 | 0.4843 |
#### Information Retrieval
* Dataset: `dim_64`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) with these parameters:
```json
{
"truncate_dim": 64
}
```
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.2719 |
| cosine_accuracy@3 | 0.485 |
| cosine_accuracy@5 | 0.5497 |
| cosine_accuracy@10 | 0.6209 |
| cosine_precision@1 | 0.2719 |
| cosine_precision@3 | 0.1617 |
| cosine_precision@5 | 0.1099 |
| cosine_precision@10 | 0.0621 |
| cosine_recall@1 | 0.2719 |
| cosine_recall@3 | 0.485 |
| cosine_recall@5 | 0.5497 |
| cosine_recall@10 | 0.6209 |
| **cosine_ndcg@10** | **0.4495** |
| cosine_mrr@10 | 0.3944 |
| cosine_map@100 | 0.4024 |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### json
* Dataset: json
* Size: 801 training samples
* Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
* Approximate statistics based on the first 801 samples:
| | anchor | positive | negative |
|:--------|:----------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|:-----------------------------------|
| type | string | string | list |
| details | <ul><li>min: 9 tokens</li><li>mean: 19.53 tokens</li><li>max: 41 tokens</li></ul> | <ul><li>min: 16 tokens</li><li>mean: 197.47 tokens</li><li>max: 410 tokens</li></ul> | <ul><li>size: 5 elements</li></ul> |
* Samples:
| anchor | positive | negative |
|:---------------------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| <code>What were the clothing details of the person identified by the special informant near the storage shed?</code> | <code>was ready to come forward as a witness. Out of necessity, we police officials searched each other to ensure none of us had any incriminating items. Subsequently, accompanied by the special informant, we proceeded to the location indicated. Upon reaching the storage shed near the crematorium ghat, the special informant pointed out a person wearing black and slate-colored jersey and jeans standing behind the wall of the storage shed. We police officials parked our motorcycles nearby, hid ourselves, and approached the back wall where the person was standing. Seeing us, the individual started running, and we police officials caught him after using necessary force. The arrested person's name and address were asked, and he stated his name as Shahvez son of Mowlai No.Mo., Kakra Khurd Thana Kotwali, Shahjahanpur, age approximately 25 years. A search of his person revealed a homemade indigenous firearm 315 bore hidden in the left pocket of his pants, with a total length of one ballist three</code> | <code>['Bablu alias Umesh Kumar Kushwaha, son of Nathoo Prasad Kushwaha, residing in Gram Pachokhar, Than Ahra, Bandha, was conducted. Opening the white plastic bag held in his right hand revealed marijuana. The weight was measured using an electronic scale borrowed from Constable Vijay Singh in the town of Attara, which showed the weight of the marijuana to be 1 kilogram 500 grams. When asked for a license to keep and sell it, he could not show it. The act of Bablu alias Umesh Kumar Kushwaha as mentioned above constitutes a punishable offense under Section 8/20 NDPS Act. Being informed of the offense, he was taken into custody at the Gram Garganpurawa police station from the Garganpurawa turnoff, about 20 steps from Garganpurawa on the river embankment road. The seized marijuana without license of 1 kilogram 500 grams was taken into police custody. A sample of 50 grams was separated from the seized marijuana, and the remaining 1.45 kilograms of marijuana and 50 grams of sample were kept', 'First Information Report: A DCM vehicle numbered UP19T5287 loaded with large animal bones and two accused under Sections 420/278 IPC, Kotwali police station, Basti district, today on the date 01.09.23, while Inspector Rajendra Yadav was touring the area with officers Akhilesh Yadav and Indrajit Paswan, along with employees of Hamaraa Chitra 18, reached near the Toll Plaza Basti when a DCM vehicle numbered UP19T5287 passed by from Gorakhpur, emitting a strong foul smell. Due to the vehicle slowing down near the toll plaza, the driver was signaled to stop on the left side of the road. The names and addresses of the driver and the person sitting with him were asked. The driver introduced himself as Mintu Kumar, son of Tellu Ram, residing in Aldi Police Station, Shamli district, age 3 NC RB (N.C.R.B) I.L.F.-I (Unified Investigation Form - I), 24 years old. The name and address of the person sitting with the driver were also asked, who introduced himself as Deshpal SIO Krishnapal, residing', '(Unified Investigation Form -1), Purva Case, Tikra Village, Asandra Police Station, Barabanki District. A search of his person revealed a plastic bottle filled with liquid substance in his right hand. Opening the cap of the bottle, we smelled a strong odor of raw liquor. When asked for a license to carry liquor, he refused to show it and requested forgiveness for his mistake. The above-mentioned person was taken into custody under Section 60(1) of the Act around 6:05 PM. During the arrest, strict adherence was given to the orders and directives of the Human Rights Commission and the Honorable Supreme Court. The bottle cap was resealed, and witnesses were sought from the public without revealing their names. The seized bottle was brought to the police station and sealed with a serial number seal. Notice of arrest will be given to the relative. The statement will be written, read, and heard before being signed by the concerned parties. Premchand Rawat, signature HC. Ali Ahmad, signature', 'police post. Upon questioning the arrested person for name and address, 3 NC R.B (N.C.R.B) I.L.F.-I (Unified Investigation Form - I) was filled, and a body search was conducted. The arrested person stated his name as Sanjay Rajput, son of Ramasare, residing in Jhinjhak Vasti Ward No. 5 Durgais Nagar Police Station Mangalpur District Dehat Kanipalwar, age approximately 35 years. During the body search, 15 units of country liquor were seized from the right hand held plastic bag. When questioned about the illegal country liquor, he stated that he buys liquor from contractors and sells it village to village for more money. When asked for a license to sell, he failed to show one and started asking for forgiveness for the mistake. The act of the arrested person falls under the ambit of Section 60 Excise Act. The arrested person was informed of the charge under Section 60 EXACT.KS and was arrested around 10:22 PM. The illegal country liquor was seized by the police. At the time of arrest,', 'First Information Report: Oral Statement - It was stated that I am SHO Sanjeev Kumar Dubey, there are 20 numbered cartridges with Dal 1443 Dalveer Singh, one each; 20 numbered cartridges with 1504 Sandeep Kumar, one each; driving a government jeep UP 85 AG 0415. Departed from Jayprakash on the day of the incident Imroz, stopped for investigation and check and ATM and the suspect vehicle and search desired criminals and investigation gang chart approved by Shri District Magistrate Mathura, Gang Leader 1. Amir Khan son Rahmat Khan residing Akeda Thana Sadar Noonu Mevata Haryana was informed that people reported in various places that Amir Khan residing Akeda Thana Sadar Noonu Mevata Haryana is part of an organized group, its members are 2. Muslim alias Hakimi son Masood alias Sunder Jamat Thana Punhana Noonu Mevata Haryana, 3. Ajriddin alias Ajru alias Ajju son Azad residing Ghagwadi Thana Kathwada Bharatpur Rajasthan, 4. Manish alias Mann son Saudan residing Kanhor Thana Pahadi']</code> |
| <code>How did the police officials manage to catch the individual after he started running?</code> | <code>was ready to come forward as a witness. Out of necessity, we police officials searched each other to ensure none of us had any incriminating items. Subsequently, accompanied by the special informant, we proceeded to the location indicated. Upon reaching the storage shed near the crematorium ghat, the special informant pointed out a person wearing black and slate-colored jersey and jeans standing behind the wall of the storage shed. We police officials parked our motorcycles nearby, hid ourselves, and approached the back wall where the person was standing. Seeing us, the individual started running, and we police officials caught him after using necessary force. The arrested person's name and address were asked, and he stated his name as Shahvez son of Mowlai No.Mo., Kakra Khurd Thana Kotwali, Shahjahanpur, age approximately 25 years. A search of his person revealed a homemade indigenous firearm 315 bore hidden in the left pocket of his pants, with a total length of one ballist three</code> | <code>['hid behind motorcycles and approached stealthily, arresting the individual named Sunil, son of Harishchandra, residing in Rudapurwa Majra, Nayagaon, Mubarakpur, Kotwali Police Station, Dehat Hardoi, approximately 36 years old. A search of his person revealed a plastic container with about 14 liters of liquid. Upon questioning, he admitted it was raw alcohol. Opening the container lid, a strong smell of raw alcohol was detected. When asked for a license to sell and possess raw alcohol, he refused to show one and requested forgiveness for his mistake. His actions constituted an offense under Section 60 ExAct. He was informed of the charges and taken into custody at 7:25 PM. The seized container was prepared with a memo, sealed, and stamped. During the arrest, orders and guidelines from the Supreme Court and National Commission were strictly followed. Some people came during the arrest but left without giving their names or addresses after being asked for witness statements. This report', 'First Information Report: This is a handwritten complaint submitted in the Hindi service, addressed to Shri Inspector Sir, Thana Motipur, Bahraich district, Sir, it is stated that the complainant Poonam, wife of Ramabhilakh, residing in Jayarampurwa, Kudwaa, Thana Motipur, Bahraich, am living there. On the date 11.01.2024, at 4 PM, Dr. Urff Ramesh, son of Bala Lodhi, residing in Kaalapurwa, Mo Nagar, Thana Ramgaon, Bahraich district, took away my daughter Nisha, aged about 14 years, daughter of Ramabhilakh. Today, my daughter told me that Dr. Urff Ramesh forcibly took her away. Due to which I have come to the police station with my daughter. Therefore, I request Mr. Sir to kindly write my case and take appropriate action against the accused. Ref No. Poonam 3 N.C.R.B (N.C.R.B) I.L.F.-1 (Unified Investigation Form -1) Complainant Poonam, wife of Ramabhilakh, Village Jayarampurwa, Kudwaa, Thana Motipur, Bahraich, Mobile No. 7052270342 Date 17.01 2024 The registration of the case was done', 'Barabanki, approximately 23 years old. From his body search, a total of 1120 rupees was recovered, including two 500 rupee notes, two 50 rupee notes, and one 20 rupee note. The second individual identified himself as Muhammad Imran, son of Shahid Ali, residing in Dhakauli Thana, Kotwali Nagar district, Barabanki, approximately 25 years old. From his body search, a total of 1030 rupees was recovered, including two 500 rupee notes, one 20 rupee note, and one 10 rupee note. The third individual identified himself as Radhe, son of Sohki Lal, residing in Dhakauli Thana, Kotwali Nagar district, Barabanki, approximately 19 years old. From his body search, a total of 720 rupees was recovered, including one 500 rupee note, two 100 rupee notes, and one 20 rupee note. The fourth individual identified himself as Sanjay, son of Kanaku, residing in Dhakauli Thana, Kotwali Nagar district, Barabanki, approximately 20 years old. From his body search, a total of 750 rupees was recovered, including one', "neck and cheeks with the intention of killing her, which put the petitioner in a critical condition and blood started flowing from her neck. He also attacked Munni Devi, a neighbor of the petitioner. Then Munni Devi started shouting loudly, so he ran downstairs with the knife. Some people downstairs caught him and called the police by dialing 112. Then the police took him to the concerned police station, and the petitioner's treatment is ongoing at Motilal Nehru Medical College where her condition has become serious. The opponent is aggressive and violent and wants to kill the petitioner. If a case is not registered against him under severe sections, an accident can happen to the petitioner at any time. The opponent is a criminal type of person from before, and his brother is also supporting him. Therefore, respectfully requested to Mr. Sir that kindly register a case against the opponent for attempted murder and kindly register the case under severe sections and kindly register the", '(First Information Report): A fake case report of 10 liters of raw liquor and arrest of one person under Section 60 of the Excise Act today on the date 22/3/24 by me, Shivraj Singh, UNI, May, UNI Shivraj Singh May, UNI Shaitan Singh HC 231 Sanjay Kumar from Haja Police Station with report No. 06 at time 1:06 PM was on duty for verification and checking of individuals, vehicles, and investigation in the area. When we reached near Samaspur Gate, we saw a sudden panic when a person holding a plastic bottle suddenly ran away quickly upon seeing us, which made us suspicious. We chased him and caught him near Samaspur Gate about 20 steps away at around 3:00 PM. Upon questioning the arrested individual, his name was Thakur Das, son of Thakur Singh RIO, police station Chatari, Vikroori, district Bulandshahr, age approximately 40 years. He was holding a white plastic can of about 10 liters in his right hand. Opening the lid of the plastic can, we smelled the odor of raw liquor. When asked for a']</code> |
| <code>What specific details were provided about the homemade indigenous firearm found on the arrested person?</code> | <code>pants, with a total length of one ballist three inches, barrel iron approximately eight inches, stock iron approximately four inches, and hammer iron approximately four inches, with wooden plates screwed on both sides, trigger hammer operational, and a pin with a cap to open and close the barrel with a spring. Additionally, a live cartridge 315 bore was found in his right pocket, with 8 mm KF written on its base. The arrested person complained of pain in the lower abdomen, stating it was due to an injury from falling in the morning. When asked for a license for the firearm and cartridge, Shahvez son of Mowlai could not show it and requested forgiveness for his mistake. Since his act falls under Section 3/25 A. Act, which is a punishable offense, the NAFKAL took him into custody at 12:40 PM, citing the reason. The seized firearm and cartridge were kept in a cloth bag, sealed, and stamped with a sample seal. Arrest and seizure memos were prepared. During the arrest and seizure, the</code> | <code>['attached with screws. The hammer is made of iron, the total length being 2 cm (1 inch). The trigger guard is made of iron with a screw for opening and closing attached to an iron plate on the other side. The seized firearm is a brass cartridge marked KFSMM on its base. This act falls under the chargeable offense under Section 3/25 A Act. At around 10:10 am, while informing about the arrest, the orders and guidelines of the Human Rights Commission and the Supreme Court were followed, and Himanshu Singh son of Veerendra Singh, residing in Paradhan, Kotwali town, Maun district, was taken into custody under Section 3/25 A Act. One motorcycle vehicle numbered UPS4 AB7344 was seized from the scene, and when asked for registration, he could not show any, which is an offense under 207 M.V. Act. A report has been filed under 207 MV Act. The seized illegal firearm and cartridge were placed in a transparent plastic box brought by Constable Ka0 Manu Kumar and sealed with a sample seal. The case', "under Section 4/25 AACT. Therefore, the accused Jitendra Pal was informed of his crime under Section 4/25 AACT and was taken into custody at 23:15 hours at the police station. The seized sword was also taken into custody, wrapped in white cloth, sealed, and a sample seal was prepared. Requests for witness statements were made to those present, but due to various reasons, everyone left without giving their names and addresses. During the arrest and seizure, strict adherence to the orders and instructions of the Supreme Court and the Human Rights Commission was maintained. The accused's arrest was informed to his family members, and the FIR was written by talking to the local residents, including Ankrit Gupta, and read out to them. Handwritten notes in Hindi and English are illegible. The notes were attested by Balveer Singh PNO 062250708 and typed by Sonu on the computer.", "and began pleading for forgiveness. At around 19:15, the apprehended individual was taken into custody under Section 60(1) of the Excise Act. A sample of 01 liter was extracted from the plastic container for examination and placed in a plastic bottle. The container and bottle were sealed with a white cloth and stamped as evidence. An arrest memo was prepared on the spot. During the seizure and arrest, strict adherence to the orders and guidelines of the Supreme Court and the Human Rights Commission was maintained. Due to the urgency of the situation, no public witness was present. The report was written and read out loud by Vivek Kumar in sufficient light from a mobile phone and torch, and all concerned parties were informed. The arrested individual's family will be notified through appropriate means upon arrival at the police station. Note: A copy of the report was given to the accused on the spot but was destroyed. Note: Copy of the original report - H.M. Ajit Kumar | 4 NC R.B", "on foot towards the mango orchard. Inside the mango orchard, we found a person standing, identified by the informant as the suspect, who then fled upon seeing us. We surrounded him and caught him inside the orchard. Upon questioning his name and address, we conducted a search, during which he identified himself as Dhiree Singh, SIO Shri Vijay Singh, aged 24, residing in Asigawan village, Kot0 Dehat Police Station, Hardoi district. During the search, we found a 12 bore gun, which he had taken out from his left pant pocket. Opening the gun, we found a live cartridge 12 bore chambered inside. The gun's description is as follows: barrel length 10 inches, wooden handle 5 inches, bolt 5 inches made of iron with wood attached and screws fitted, trigger below the handle, hammer and copper pin above the handle, a round iron plate attached to open the barrel, pulling back the hammer activates the gun, the trigger and gun are operational, a 12 bore live cartridge was removed from the chamber,", "it by showing himself using a torchlight. All passersby witnessed the incident and were asked to testify if needed, but they left due to fear. The complaint was written, read, and signed by both the complainant and the accused on the spot. A carbon copy of the complaint was given to the accused, and the arrest notification was provided to the accused's relatives through appropriate means. Signature C. Dharmpal, HEO Vinod Kumar, Police Station Behta Mujawar, Unnao District, Mark Thumbprint Accused Denoo. Date 02.06.2021 Note: The complaint was typed word-for-word on the computer by me, HEO. 4 NC RB (NCRB) ILF-1 (Unified Investigation Form - 1)"]</code> |
* Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
```json
{
"scale": 20.0,
"similarity_fct": "cos_sim"
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: epoch
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 16
- `gradient_accumulation_steps`: 8
- `learning_rate`: 2e-05
- `num_train_epochs`: 50
- `lr_scheduler_type`: cosine
- `warmup_ratio`: 0.1
- `bf16`: True
- `tf32`: True
- `load_best_model_at_end`: True
- `optim`: adamw_torch_fused
- `batch_sampler`: no_duplicates
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: epoch
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 8
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 2e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 50
- `max_steps`: -1
- `lr_scheduler_type`: cosine
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.1
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: True
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: True
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: True
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch_fused
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: None
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `include_for_metrics`: []
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `use_liger_kernel`: False
- `eval_use_gather_object`: False
- `average_tokens_across_devices`: False
- `prompts`: None
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss | dim_1024_cosine_ndcg@10 | dim_768_cosine_ndcg@10 | dim_512_cosine_ndcg@10 | dim_256_cosine_ndcg@10 | dim_128_cosine_ndcg@10 | dim_64_cosine_ndcg@10 |
|:--------:|:------:|:-------------:|:-----------------------:|:----------------------:|:----------------------:|:----------------------:|:----------------------:|:---------------------:|
| 1.0 | 4 | - | 0.5307 | 0.5307 | 0.5140 | 0.4598 | 0.3836 | 0.2600 |
| 2.0 | 8 | - | 0.5500 | 0.5500 | 0.5329 | 0.4736 | 0.3953 | 0.2692 |
| 2.6154 | 10 | 21.2344 | - | - | - | - | - | - |
| 3.0 | 12 | - | 0.5777 | 0.5777 | 0.5644 | 0.5024 | 0.4182 | 0.2931 |
| 4.0 | 16 | - | 0.6006 | 0.6006 | 0.5897 | 0.5327 | 0.4507 | 0.3240 |
| 5.0 | 20 | 15.3903 | 0.6073 | 0.6073 | 0.5992 | 0.5528 | 0.4884 | 0.3596 |
| 6.0 | 24 | - | 0.5985 | 0.5985 | 0.5978 | 0.5574 | 0.5060 | 0.3928 |
| 7.0 | 28 | - | 0.5997 | 0.5997 | 0.5919 | 0.5598 | 0.5199 | 0.4097 |
| 7.6154 | 30 | 11.3886 | - | - | - | - | - | - |
| 8.0 | 32 | - | 0.5996 | 0.5996 | 0.5934 | 0.5659 | 0.5254 | 0.4136 |
| 9.0 | 36 | - | 0.6068 | 0.6068 | 0.5969 | 0.5743 | 0.5326 | 0.4163 |
| 10.0 | 40 | 6.435 | 0.6084 | 0.6084 | 0.6003 | 0.5792 | 0.5376 | 0.4281 |
| 11.0 | 44 | - | 0.6133 | 0.6133 | 0.6087 | 0.5801 | 0.5446 | 0.4370 |
| 12.0 | 48 | - | 0.6207 | 0.6207 | 0.6102 | 0.5898 | 0.5459 | 0.4448 |
| 12.6154 | 50 | 4.2559 | - | - | - | - | - | - |
| **13.0** | **52** | **-** | **0.6285** | **0.6285** | **0.6171** | **0.5947** | **0.5488** | **0.4461** |
| 14.0 | 56 | - | 0.6236 | 0.6236 | 0.6111 | 0.5925 | 0.5483 | 0.4556 |
| 15.0 | 60 | 2.2983 | 0.6216 | 0.6216 | 0.6144 | 0.5899 | 0.5492 | 0.4556 |
| 16.0 | 64 | - | 0.6177 | 0.6177 | 0.6116 | 0.5883 | 0.5504 | 0.4552 |
| 17.0 | 68 | - | 0.6178 | 0.6178 | 0.6095 | 0.5866 | 0.5433 | 0.4508 |
| 17.6154 | 70 | 1.7678 | - | - | - | - | - | - |
| 18.0 | 72 | - | 0.6127 | 0.6127 | 0.6030 | 0.5812 | 0.5392 | 0.4491 |
| 19.0 | 76 | - | 0.6071 | 0.6071 | 0.6018 | 0.5805 | 0.5381 | 0.4479 |
| 20.0 | 80 | 1.1051 | 0.6082 | 0.6082 | 0.5988 | 0.5781 | 0.5338 | 0.4438 |
| 21.0 | 84 | - | 0.6085 | 0.6085 | 0.6001 | 0.5788 | 0.5340 | 0.4448 |
| 22.0 | 88 | - | 0.6119 | 0.6119 | 0.6050 | 0.5826 | 0.5396 | 0.4483 |
| 22.6154 | 90 | 0.8751 | - | - | - | - | - | - |
| 23.0 | 92 | - | 0.6143 | 0.6143 | 0.6088 | 0.5858 | 0.5413 | 0.4467 |
| 24.0 | 96 | - | 0.6156 | 0.6156 | 0.6101 | 0.5886 | 0.5427 | 0.4510 |
| 25.0 | 100 | 0.6431 | 0.6157 | 0.6157 | 0.6103 | 0.5882 | 0.5464 | 0.4530 |
| 26.0 | 104 | - | 0.6100 | 0.6100 | 0.6041 | 0.5820 | 0.5407 | 0.4479 |
| 27.0 | 108 | - | 0.6074 | 0.6074 | 0.6001 | 0.5814 | 0.5377 | 0.4472 |
| 27.6154 | 110 | 0.6123 | - | - | - | - | - | - |
| 28.0 | 112 | - | 0.6069 | 0.6069 | 0.6026 | 0.5797 | 0.5362 | 0.4481 |
| 29.0 | 116 | - | 0.6068 | 0.6068 | 0.6006 | 0.5806 | 0.5355 | 0.4447 |
| 30.0 | 120 | 0.4621 | 0.6085 | 0.6085 | 0.5993 | 0.5809 | 0.5348 | 0.4441 |
| 31.0 | 124 | - | 0.6076 | 0.6076 | 0.5980 | 0.5791 | 0.5342 | 0.4464 |
| 32.0 | 128 | - | 0.6081 | 0.6081 | 0.5946 | 0.5781 | 0.5305 | 0.4424 |
| 32.6154 | 130 | 0.4917 | - | - | - | - | - | - |
| 33.0 | 132 | - | 0.6057 | 0.6057 | 0.5966 | 0.5746 | 0.5324 | 0.4436 |
| 34.0 | 136 | - | 0.6048 | 0.6048 | 0.5939 | 0.5756 | 0.5307 | 0.4458 |
| 35.0 | 140 | 0.3875 | 0.6059 | 0.6059 | 0.5980 | 0.5776 | 0.5336 | 0.4475 |
| 36.0 | 144 | - | 0.6061 | 0.6061 | 0.5966 | 0.5763 | 0.5347 | 0.4465 |
| 37.0 | 148 | - | 0.6079 | 0.6079 | 0.5937 | 0.5761 | 0.5316 | 0.4444 |
| 37.6154 | 150 | 0.4292 | - | - | - | - | - | - |
| 38.0 | 152 | - | 0.6063 | 0.6063 | 0.5927 | 0.5752 | 0.5298 | 0.4427 |
| 39.0 | 156 | - | 0.6078 | 0.6078 | 0.5956 | 0.5756 | 0.5304 | 0.4436 |
| 40.0 | 160 | 0.3469 | 0.6064 | 0.6064 | 0.5950 | 0.5748 | 0.5314 | 0.4426 |
| 41.0 | 164 | - | 0.6092 | 0.6092 | 0.5982 | 0.5782 | 0.5305 | 0.4450 |
| 42.0 | 168 | - | 0.6084 | 0.6084 | 0.5988 | 0.5783 | 0.5330 | 0.4466 |
| 42.6154 | 170 | 0.3876 | - | - | - | - | - | - |
| 43.0 | 172 | - | 0.6084 | 0.6084 | 0.6014 | 0.5791 | 0.5360 | 0.4495 |
| 44.0 | 176 | - | 0.6095 | 0.6095 | 0.6008 | 0.5800 | 0.5379 | 0.4489 |
| 45.0 | 180 | 0.358 | 0.6102 | 0.6102 | 0.6019 | 0.5809 | 0.5371 | 0.4486 |
| 46.0 | 184 | - | 0.6098 | 0.6098 | 0.6009 | 0.5783 | 0.5391 | 0.4479 |
| 47.0 | 188 | - | 0.6083 | 0.6083 | 0.6009 | 0.5794 | 0.5369 | 0.4487 |
| 47.6154 | 190 | 0.3588 | - | - | - | - | - | - |
| 48.0 | 192 | - | 0.6087 | 0.6087 | 0.6031 | 0.5799 | 0.5358 | 0.4480 |
| 49.0 | 196 | - | 0.6100 | 0.6100 | 0.6020 | 0.5802 | 0.5362 | 0.4482 |
| 50.0 | 200 | 0.2932 | 0.6082 | 0.6082 | 0.6014 | 0.5814 | 0.5366 | 0.4495 |
* The bold row denotes the saved checkpoint.
### Framework Versions
- Python: 3.10.12
- Sentence Transformers: 4.1.0
- Transformers: 4.52.2
- PyTorch: 2.6.0+cu124
- Accelerate: 1.7.0
- Datasets: 3.6.0
- Tokenizers: 0.21.1
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### MultipleNegativesRankingLoss
```bibtex
@misc{henderson2017efficient,
title={Efficient Natural Language Response Suggestion for Smart Reply},
author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
year={2017},
eprint={1705.00652},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->